A Graph-to-Sequence Model for AMR-to-Text Generation
This addresses text generation from semantic graphs for NLP applications, but it is incremental as it builds on existing graph encoding approaches.
The paper tackled the problem of AMR-to-text generation, where current methods lose structural information from graphs, and introduced a neural graph-to-sequence model that achieved superior results on a standard benchmark.
The problem of AMR-to-text generation is to recover a text representing the same meaning as an input AMR graph. The current state-of-the-art method uses a sequence-to-sequence model, leveraging LSTM for encoding a linearized AMR structure. Although being able to model non-local semantic information, a sequence LSTM can lose information from the AMR graph structure, and thus faces challenges with large graphs, which result in long sequences. We introduce a neural graph-to-sequence model, using a novel LSTM structure for directly encoding graph-level semantics. On a standard benchmark, our model shows superior results to existing methods in the literature.